The Fitch Paradox and the Factivity of Knowability

The article examines the question of whether knowability is factive. Most researchers admit that knowledge is factive, which is due to the standard definition of knowledge as true justified belief. But the question of the factivity of knowability is much more complex. This is especially evident in the context of debates on the Fitch paradox. The paradox shows that the straightforward formalization of ‘p is knowable’ as ‘it is possible that p is known’ is incorrect because it readily leads to a contradiction (granted some intuitively appealing principles). Those who want to preserve the principle of knowability (according to which every truth is knowable) attempt to elaborate alternative formalizations of knowability, free from paradoxical implications. Among them are Dorothy Edgington and Michael Fara. Edgington defines p’s knowability as the possibility to know that p is true in the actual world, which renders knowability factive. Timothy Williamson’s criticism on her proposal shows that it has some substantial flaws. Fara wants to amend Edgington’s proposal preserving the factivity of knowability. He defines knowability of a proposition p as there actually being an agent who is capable to know that p is actually true. This definition also implies that knowability is factive. I examine Fara’s proposal and show that it is essentially incomplete because of two reasons. First, Fara does not provide syntactic rules for the operator ‘C_x’ that he introduces to symbolize the agents’ capacity to have knowledge. Secondly, I show that the standard modal semantics cannot be used in interpreting his definition of knowability, whereas he does not propose any nonstandard semantics for it. I argue that the flaws of both theories are due to the fact that they treat knowability as factive. This motivates my hypothesis that a concept of knowability free from factivity is needed to solve Fitch’s paradox.

Keywords

Authors

References